Skip navigation

Home | About | Contact

Digital Library > Journals > JILR > Volume 23, Issue 3 >
Login or register for free to remove ads.

Template Authoring Environment for the Automatic Generation of Narrative Content

New Search
New Search
Print Abstract
Print Abstract
E-mail Abstract
E-mail Abstract
Full Text
Full Text
Add To Collection
Save to My Collections
Export Citation
Export Citation

Caropreso, M.F., Inkpen, D., Keshtkar, F. & Khan, S. (2012). Template Authoring Environment for the Automatic Generation of Narrative Content. Journal of Interactive Learning Research, 23(3), 227-249. Chesapeake, VA: AACE.
Retrieved from http://www.editlib.org/p/37515.

OpenURL Link

Journal Information

JILR

Journal of Interactive Learning Research
ISSN 1093-023X
Volume 23, Issue 3, July 2012
Association for the Advancement of Computing in Education (AACE)  Chesapeake, VA

More Information on JILR

Table of Contents


Authors

Maria Fernanda Caropreso, Diana Inkpen, Fazel Keshtkar, Shahzad Khan, University of Ottawa, Canada

Abstract

Natural Language Generation (NLG) systems can make data accessible in an easily digestible textual form; but using such systems requires sophisticated linguistic and sometimes even programming knowledge. We have designed and implemented an environment for creating and modifying NLG templates that requires no programming knowledge, and can operate with a minimum of linguistic knowledge. It allows specifying templates with any number of variables and dependencies between them. It internally uses an existing sentence realization NLG tool in order to provide the linguistic background knowledge. We tested the performance and usability of our system in the context of interactive simulation games. We incrementally improved our system in order to obtain all the capabilities needed to reproduce all the sentences and templates manually created for already existing games. We trained the users and measured their satisfaction with the system by comparing the results of writing new games’ narrative content manually vs. using our system. In general, the use of the system made the task faster, more enjoyable, and less prone to errors.

Also Read

Tags

Comments & Discussion

Comment on the paper above. You must be registered to participate. Registration is free.




Feedback and Suggestions please email info@editlib.org.